How our search algorithm works

Overview

Search on Liners uses a hybrid approach. We combine traditional keyword matching with semantic search so users can find products by name, category, use case, or natural-language need.

Keyword search is useful when someone knows the exact product or phrase. Semantic search helps when someone describes a problem, such as "send money abroad" or "tools for restaurant payments," and the best product pages may not use those exact words.

How It Works

When you type a query, two things happen in parallel:

1. Keyword search: We match your query against product names, taglines, descriptions, and other listing text.
2. Semantic search: We compare the meaning of your query with product embeddings to find products that may be relevant even when the wording differs.

Results from both are merged and ranked:

• Exact name matches always float to the top.
• Semantic matches are ranked by how similar they are to your query.
• Keyword-only matches fill the remaining slots.
• Duplicates are automatically removed.

Example

Searching for "send money abroad" will surface products related to international transfers and remittances, even if they don't contain those exact words anywhere in their listing.

Search Quotas

Semantic search costs more than keyword search, so Liners applies usage limits to keep search reliable and sustainable. Limits may vary by account state and operational needs.

When you've used up your quota:

• Search automatically falls back to keyword-only results, still fully functional, just without the semantic layer.
• You'll see a message letting you know, with a suggestion to sign in (for guests) or that you've hit your daily cap (for logged-in users).
• Your quota resets every 24 hours.

Fallback Behaviour

If semantic search is unavailable, search falls back to keyword-only results. The goal is that users can still search the directory even when the AI-assisted layer is unavailable.