Vector Search (Fallback)

Prev Next

By configuring these settings, you can ensure a smoother and more intelligent search experience for your users, especially in scenarios where keyword-based search comes up empty.

Vector Search Fallback

The Vector Search 0-Hit Fallback feature enhances the search experience by eliminating zero-result queries. It enables support for conversational and natural language search, including long-tail queries that may not directly match product descriptions.

Key Capabilities:

  • No More Zero Results: Ensures users receive relevant results, even when exact keyword matches are unavailable.

  • Context-Aware Understanding: Utilizes vector embeddings (also referred to as embedding models or language models) to grasp the intent and context behind user queries.

  • Natural Language Support: Allows users to search using everyday language, making the search experience more intuitive and flexible.

By understanding the meaning behind queries instead of relying solely on exact keyword matches, Vector Search 0-Hit Fallback returns more relevant and meaningful results—helping users find what they’re looking for, even when terminology differs.

Note: This feature leverages vector-based language understanding technology but does not rely on general-purpose AI or large language models (LLMs). Your data stays on our servers, locally.

Vector Search Settings

The Vector Search Settings Page allows you to configure what happens when a keyword search returns no results (also known as a "0-hit" search). With this feature, you can enable a fallback to vector-based search, helping users find relevant results even when traditional keyword matching fails.

Availability

This settings page is only visible if the Vector Search module is enabled for your account.


Settings

Vector Search Active

  • Description: Enables or disables vector search fallback behavior.

  • Default: Off

  • When enabled: The system will use vector search to retrieve results when a keyword search returns no matches.

Limit

  • Description: Sets the maximum number of results to return from a vector search to prevent using up your account limit too rapidly.

  • Default: 1000

  • Type: Integer