July 7, 2026
8 min read
Ten features that separate good from generic
1. Natural language query understanding
Handles “gift for hiker”, “vegan friendly”, “oversized” without keyword tricks.
2. Real-time stock and price awareness
Stock changes reflected in ranking within minutes, not hours.
3. Typo tolerance and synonyms
Including generative synonyms that don’t require manual configuration.
4. Dynamic faceted navigation
Facets adapt to category, query, and result set. Zero-result suppression on filters.
5. Marketer-operable ranking rules
Push margin-tier or seasonal bias without engineering tickets.
6. Personalisation by segment and behaviour
Same query, different ranking by audience or individual.
7. Out-of-stock and zero-result fallback
Never a dead-end page.
8. Search-page recommendations
“You might also like” blocks on the search results page, using the same engine.
9. Per-query attribution
Revenue measured per query with clean holdout testing.
10. Native integration with your ecommerce platform
App-store install, not custom integration.
Platforms strong across most features
Clerk.io
All ten features, with optional built-in AI agent handling configuration. Native integrations with Shopify, BigCommerce, Magento, WooCommerce and Prestashop.
Algolia
Strong on most features, developer-led configuration. Trade-offs on the Algolia alternative page.
Klevu
Marketer-operable across most features, Shopify and BigCommerce-strong. Trade-offs on the Klevu alternative page.
Bloomreach Discovery
Enterprise scope with deep features and CDP integration. Trade-offs on the Bloomreach alternative page.
Searchspring
Strong manual merchandising control with most features available.
How to evaluate them in a demo
- Bring 50 of your busiest products with realistic stock state
- Test natural language queries against your catalogue, not the vendor’s
- Push a ranking rule live and verify it took effect
- Ask for per-query attribution methodology in writing
- Get pricing including renewal escalator
How to select a vendor for advanced ecommerce search
Do not turn the ten features into a checkbox race. Connect each one to a real query, merchant workflow, data source, and measurable outcome.
| Vendor type | Representative providers | Capabilities to demonstrate | Workload to clarify |
|---|---|---|---|
| Commerce search and discovery platform | Clerk, Klevu, Bloomreach, Searchspring | Relevance, facets, merchant rules, recommendations, analytics, and commerce connectors | Feed setup, storefront changes, markets, training, and contracted modules |
| API-led hosted search | Algolia | Keyword and semantic retrieval, UI libraries, facets, rules, events, analytics, and merchandising | Frontend ownership, index design, event implementation, usage, and relevance operations |
| Search infrastructure | Elastic or Solr | Retrieval, ranking, schemas, facets, scaling, and deployment controls | Commerce data model, merchant interface, storefront, monitoring, security, and upgrades |
The vendor names are examples, not a claim that packages or implementations are equivalent. Product scope and plan access change. Ask for a dated written response against your own requirements.
Use seven proof tests
- AI relevance: Grade natural-language and vague-intent queries from real search logs.
- Facets: Apply several filters on desktop and mobile, then test counts and zero-result prevention.
- Merchandising: Have a merchant pin, boost, demote, exclude, schedule, preview, and undo a rule.
- Analytics: Trace a query through a click, cart addition, order, and revenue report.
- Personalization: Repeat one query with different session behavior and inspect the result change.
- Integration: Change stock, price, and a product attribute, then record when search updates.
- Scale: Repeat the test with production-like catalog size, variants, traffic, markets, and languages.
Clerk Intelligent Search works with Recommendations, Audience, Email, and Chat. Clerk’s approved position is no developer work required, no cold-start issues, accurate results immediately, and completely cookieless and free from storing customer data.
For a real storefront example, see GrejFreak’s customer story, which shows search and recommendation placements across a specialist outdoor catalog.
Use the ROI calculator to compare expected commercial return with software, implementation, and operating cost.
TL;DR
- Ten features separate good ecommerce search from generic: query understanding, real-time data, typo and synonym handling, dynamic facets, marketer rules, personalisation, zero-result handling, search-page recs, attribution, native integration.
- Clerk.io, Algolia, Klevu, Bloomreach Discovery and Searchspring are the most commonly evaluated platforms across these features.
- Select with live proof tests for relevance, facets, merchandising, analytics, personalization, integration, and scale.
