Skip to content

Headless Commerce Search — API-First Architecture with AACSearch

Alex Chibilyaev

5/22/2026

#headless#commerce#search#AACSearch#architecture
Headless Commerce Search — API-First Architecture with AACSearch

The Monolith Trap in Modern Commerce

Most "headless" commerce setups aren't truly decoupled. The storefront is a SPA, the CMS runs on a separate instance, the cart lives in a backend API — but search is still a monolith buried in the legacy platform. You query the CMS directly, hit its MySQL fulltext index, and pray it returns something relevant.

That worked when product catalogs had 500 SKUs. It breaks at 5,000. And at 50,000 it's completely unusable.

The fix is obvious: decouple search the same way you decoupled everything else. But building a dedicated search service from scratch means choosing between a complex self-hosted engine (traditional search backends, other search engines, AACSearch) and an expensive managed SaaS that doesn't speak your stack's language.

What API-First Search Actually Means

API-first search means the search engine is an independent service that your frontend communicates with directly — no CMS middleware, no backend proxy, no database bottleneck. The architecture looks like this:

  • Index layer: Product data is synced from your CMS or commerce platform into a dedicated search index
  • Query layer: The storefront sends search requests straight to the search API via REST or SDK
  • UI layer: A search widget or custom component renders results, facets, autocomplete, and typo-tolerant suggestions

Each layer is independently deployable, scalable, and replaceable. You can switch CMS platforms without rewriting your search logic. You can A/B test ranking algorithms without touching your storefront code. That's the promise of headless search.

AACSearch as the Engine Layer

Under the hood, AACsearch is powered by AACSearch — an open-source search engine that's faster and simpler than traditional search backends for most commerce use cases. AACSearch delivers search results in milliseconds, supports typo tolerance out of the box, and handles faceted navigation, sorting, and geolocation filtering without a PhD in query DSLs.

What makes AACSearch particularly well-suited for headless commerce:

  • Instant search: Typed queries return in under 50ms — no debounce required
  • Built-in faceting: Category filters, price ranges, attribute drill-downs with zero custom infrastructure
  • Automatic typo tolerance: "Nike" matches "Nikee" and "Nik" without configuration
  • REST-first design: Every operation is a straightforward HTTP request — no custom protocol, no binary clients

Where AACsearch Abstracts the Complexity

Running AACSearch yourself means managing schema migrations, connection pooling, index replication, and delta sync schedules. That's workable when search is your product; it's a distraction when you're running a store.

AACsearch wraps AACSearch with managed infrastructure that's commerce-aware:

  • CMS connectors for Shopify, WooCommerce, PrestaShop, and Bitrix — each one maps product fields, categories, variants, and inventory automatically
  • Automatic delta syncs — when a product price changes or inventory updates, the connector picks it up and reindexes within minutes
  • Schema inference — the connector reads your product catalog and builds the AACSearch schema without manual mapping
  • A REST API and SDKs — Node.js, Python, PHP, Ruby — so you can interact with search programmatically without writing raw HTTP calls

You get the performance of a purpose-built search engine with the operational cost of a third-party integration.

Widget SDK: UI Without the Boilerplate

The search UI is the part your customers actually see, so it needs to be fast, accessible, and on-brand. AACsearch ships a widget SDK that works with React, Vue, Angular, and vanilla JavaScript.

Drop in a pre-built search box that handles autocomplete, faceted filters, and instant results — or use the SDK primitives to build a fully custom UI. The widget talks directly to the AACSearch API via your tenants' API keys, so there's no backend round-trip between the browser and your server. Every millisecond matters when a shopper types a query, and cutting out the middleware shaves 100-200ms off every search.

Putting It All Together

Here's what a headless search stack looks like with AACsearch:

  1. Your CMS (Shopify, WooCommerce, PrestaShop, Bitrix) pushes product data via the AACsearch connector
  2. AACsearch manages the AACSearch index — schema, sync, and scaling
  3. Your storefront (Next.js, Nuxt, SPA, or static site) loads the widget SDK
  4. Shopper queries go directly from browser → AACSearch API → rendered results
  5. The AACsearch dashboard surfaces analytics — top queries, zero-result searches, click-through rates

No infrastructure. No middleware. No DevOps overhead. Just fast, relevant search that scales as your catalog grows.

If you're running a headless commerce stack and search is still a bottleneck — try AACsearch free. No credit card required.